Expression and Manifestation

Random musings on the two confusing/controversial FRBR Group 1 entities, which started out as a comment on another forum, but I figure, hey, why not put it here.

Both the “manifestation” and “expression” entities together basically make up the traditional/colloquial idea of ‘edition’. If two different editions have exactly the same (or same enough for contextual purposes) text or content, then they are two manifestations of the same expression. For instance, the 1972 edition and a 1981 edition or whatever. Might have different ISBNs even. But contain the same text, then different manifestations of the same expression.

If, on the other hand, a new edition is seriously revised, then it’s also a new expression, not just a new manifestation.
You can come up with all sorts of grey areas easily, and the FRBR model does not try to precisely instruct you how to decide these grey areas. That’s the role of ‘Guidance’, not ‘model’. The FRBR Model decides it’s useful/necessary to have this distinction, it’s up to implementing communities to specify a useful way to make it as a practical matter.  Whether there’s a need for expression AND manifestation, instead of just one ‘edition’ entity is one of the more controversial parts of the FRBR ontology. I think they chose right though.

I also like thinking of this with Elaine Svenonius’ set theory approach, so from the bottom up:

An item, is an actual individual concrete book in your hand.

A manifestation is the set of all items that are identical (or close enough for our purposes) in _physical form_ as well as content.

An expression is the set of all manifestations that are identical in _textual_ or _information_ content. (or close enough for our purposes; an archeologist would consider the coffee stain on the back to be distinguishing information content; we in libraries do not even consider it distinguishing physical form, it’s still part of the same manifestation as the items without coffee stains).

And a work is the set of all expressions that well, consist of the same intellectual work. This is definitely a cultural concept, but it’s one we have and find useful. We consider the audio book version of a book to be the same “book“,  just a different version. That’s work.

(If the audio book is  abridged it’s a different expression of the work; what if it’s the full text, but read outloud? I’m actually not sure if it should be considered the same or a different expression, it’s kind of arbitrary really, and up to the community to decide.)


7 thoughts on “Expression and Manifestation”

  1. Jonathan

    I think that’s an great summary of Gp 1 entities. But the nagging worry I have is that the FRBR entities are based on a print publishing view of the world. Are these entities equally applicable to the non-print world? If we were just looking at digital resources, would we carve up the digital world into these same entities? I don’t think so. Look at Dublin Core, for example. The concepts of work, expression, etc. didn’t exist in the DC universe until the recent eprints application profile. It could be a mistake (if nothing else, a waste of time) to impose a print universe ontology on a digital universe.

  2. I think that’s a good point. The FRBR model is a formalization of traditional library practie, taking ambiguous and implicit shared mental models, and making them rigorous and explicit. It is definitely a formalization of library cataloging tradition, NOT a re-thinking from the ground up in the digital world.

    So IS it appropriate for the digital world? It’s hard to say. The fact that DC didn’t have it until recently is, to my mind, no evidence either way—the original DC was intended as a grossly more simple descriptoin than we are used to, too simple to need it. When the DC community started trying to flesh out more sophisticated and complete metadata models, that they ended up using (a version of) FRBR—this is fact suggests that FRBR _is_ useful in the digital world.

    But it’s just a suggestion, not any kind of proof. Truth is, I don’t think the analysis or the empirical testing in the crucible of practice is or has taken place to answer the question of how FRBR fits the digital world.

    Thing is, we need an explicit and formalized Domain Model. FRBR is our first attempt to do that. We need one that works for the traditional print world (which is of course, lest we forget, still very much with us and a large part of what we do) AND for the digital world. Is FRBR perfect, is it optimal? Surely not. But it’s where we’re starting, it’s the only step we’ve made so far. More work needs to be done to develop it further, both intellectual and empirical work. We don’t do enough work like that. Somehow our field needs to become an innovative intellectual field with a significant research component, which we don’t have much of now.

    But I think the FRBR model is a great start. it is our community beginning to formalize our conceptual model, and that’s something that needs to be done. Even if FRBR makes more sense for print than for digital, that’s an accomplishment in itself, the print world is complicated and messy enough already (it took 100 years of experience to figure out how to model it)–and the digital world even more so. But I’m not convinced the Group 1 Entity model won’t work for the digital world too.

    If you were starting from scratch, what conceptual ontology would you use for the digital world?

  3. That’s a good question! I think the answer is: it depends on what digital collection we’re looking at. Generally, I’d say the web is a much flatter universe in FRBR terms. In the area I’m currently working in (digital learning resources) I don’t often see multiple resources with essentially the same content that would necessitate a separate work entity, let alone expression or manifestation. Items are ephemeral and new versions replace old ones, which disappear completely. Other resources like websites and blogs are unique — and are continually changing in content. The whole publishing model is different. Does is even make sense to talk about a publisher in the digital world? A collection of digitised books is obviously closer to the print situation, ditto the e-prints. But having said that, after my quick look at the e-prints work I remember thinking: this seems to be overkill; are all these levels really necessary? You _can_ treat any digital resource as if it were a book — but it feels wrong. It’s like those early European settlers who came to Australia but persisted in wearing their heavy winter clothes in the new climate.

    I think the ‘metadata communities’ (I mean, apart from cataloguers) are yet to seriously tackle the issues of what entities we are describing. The std I work with most, Learning Object Metadata, has one entity, “learning object”, which is defined in such a way that it can be anything in the universe that someone decides to call a learning object. BUT, its attributes indicate the writers of the standard had a particular, but frustratingly never explicitly specified, type of object in mind. Likewise, DC has (or had) its ‘document-like object’.

    So, in short, I don’t have a good answer :-) but I think deciding and declaring upfront what entities are in the collection you are describing is important.

  4. It strikes me that the aggregation relationship is much more important for a large proportion of online resources, than the workset relationship. This is something that matters for a significant (but lesser) proportion of physical resources too (from anthologies to serials!), and the FRBR model hasn’t quite figured out how to model it appropriately yet. I believe there is a working group on aggregation still working.

    The series/serial relationship is also potentially significant in electronic resources, another area where the FRBR model could use some work based on both analysis and research.

    This sort of modelling needs to happen from an inductive process as well as a deductive process. By which I mean to say, people working on actual applications need to more often do what Irvin is suggesting, be more rigorous with the domain model of their particular application. From all that, we can try to abstract a general model we can all share.

  5. In my mind, the e-prints application profile’s implementation of the FRBR model solves a very particular problem, in that an electronic pre-print is often a separate _expression_ of a particular work of scholarship, whereas, in other cases a post-print may be a different manifestation of a print article (given the difference in format), and may or may not be the the same expression as the pre-print. When citing a resource, it’s a challenge to figure out exactly what the relationships between versions of things are. Especially when we’re dealing with digital things that have print analogs. I suspect this problem will only be compounded as the OAI-ORE work continues; when citing a resource, it will be important to be able to determine whether or not a copy hosted elsewhere is the same expression.

    However, even in the purely born-digital world, I can think of countless ramifications of the FRBR model. Irvin mentions blogs, and that the whole publishing model is different. I won’t address blogs, except to reiterate Jonathon’s point about the FRBR Aggregates Working Group.

    I will say something about wikis, though, using Wikipedia as an example. Take the Wikipedia entry on FRBR, which I’d at this time define as a work, even if it is part of something else:

    Well, I can easily point you to another expression of the same work:

    This happens to be the first expression (tiny URL used to not cause problems with this blog’s formatting). There are approximately 500 different “versions”, – though this post emphasizes the question of whether they’re expressions or manifestations. Then, there’s
    Wikipedia’s use of the GFDL, which leads things like’s reuse of Wikipedia content. I think these are probably manifestations?

    To an extent, issues of expression/manifestation are present throughout the web. Perfect examples include the Internet Archives’ Wayback Machine, and the Google Cache. Google’s cache gets even more interesting when looking at caches of Wikipedia articles, and other versioned content, but I’ve droned on long enough.

  6. You’re right, Corey. I was forgetting all those potential expressions/manifestations. Not to mention the versions (items?) cached on my machine when I follow a link to those pages.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s